An automatic implantable cardioverter-defibrillator (AICD) is placed in patients efficacous. Following-up in th outpatient CRC. To best effectiveness of an AA drug, a subgroup of AICD patients are randomly allocated to amiodarone or placebo. Endpoints include AICD firing or death. To test whether poor adjustment to AICD placement can be predicted before operation, psychosocial testing using SCL-90R and SIP instruments is done pre-operatively and at 4 and 12 months after operation.